Pattern of morphological variation among five wild populations of endangered Tor putitora (Hamilton, 1882) from India using truss network system

Abstract

Tor putitora (Hamilton, 1822) is a commercially important freshwater cyprinid fish distributed widely in India and adjacent countries. Despite its extensive distribution the stock structure of the species is not well known. The present study investigated intraspecific variation of wild populations of this species based on the analysis of morphometric distances of body parts using the truss network system constructed from digitized images. A total number of 210 specimens of the species were collected from five geographically separate rivers of India viz. River Beas, Ravi, Chenab, Kosi and Jia bhoreli. A truss network was constructed based on 12 landmarks plotted on the digital image of fish to form a total of 30 distance variables using tpsDig2 and PAST software platforms. The transformed truss measurements were subjected to principal component analysis, discriminant function analysis and univariate analysis of variance. Principal component analysis generated eight components explaining 86.5 % of total intraspecific variance in samples. Step-wise discriminant function analysis retained four variables that significantly discriminated the populations. Using these variables, 100 % of the original groups were classified into their correct samples and similar percentage of cross validated groups omitting one procedure was classified into their correct samples. The occurrence of multi-stock structure may be due to differences in physical and ecological parameters of the five rivers.